betaFeedback

WorksPhysics and Astronomy

Wafer-to-Wafer Bonding: Part: I – The Coupled Physics Problem and the 2D Finite Element Implementation

Abstract

Wafer-to-wafer (WxW) bonding is a key enabler for three-dimensional integration, including hybrid bonding for fine-pitch Cu–Cu interconnects. During bonding, wafer deformation and the air entrapped between the wafers interact through a strongly coupled, time-dependent fluid–structure interaction (FSI) that can produce non-intuitive bonding dynamics and process sensitivities. This paper develops a mathematically consistent reduced-order model for WxW bonding by deriving a Kirchhoff–Love plate equation for wafer bending from three-dimensional linear elasticity and coupling it to a Reynolds lubrication equation for the inter-wafer air film. The resulting nonlinear plate–Reynolds system is discretized and solved monolithically in the high-performance FEniCSx framework using a C0C^0 interior-penalty formulation for the fourth-order plate operator, standard continuous Galerkin discretization for the pressure field, implicit time integration, and a Newton solver with automatic differentiation. Simulations reproduce experimentally reported wafer displacement histories for multiple initial gaps and verify force equilibrium at the bond front, where the Reynolds pressure acts as an effective contact reaction. Parametric studies reveal nonlinear, and in some cases non-monotonic, sensitivities of bonding-front kinetics to the initial gap, air viscosity, and interfacial energy, providing actionable trends for process optimization.

Introduction

Wafer-to-Wafer (WxW) bonding has emerged as a cornerstone technology for next-generation semiconductor packaging, enabling three-dimensional integrated circuit (3D-IC) stacking, High Bandwidth Memory (HBM), MEMS fabrication, and photonic device integration [1, 2], [3]. Unlike traditional bump-based stacking, which introduces inter-die gaps of approximately 30 μ\mum and is limited to interconnect pitches above 10 μ\mum, (hybrid) WxW bonding achieves direct Cu-to-Cu connections with pitches below 10 μ\mum, enabling substantially higher I/O density, reduced package dimensions, lower electrical resistance and superior thermal performance [1], [4]. These attributes translate directly into higher bandwidth, improved signal integrity, reduced power consumption, and enhanced heat dissipation, all critical requirements for enabling miniaturization of packages and modern HBM architectures. Despite this transformative potential, WxW bonding involves high process sensitivities, such as - submicrometer particulate contamination can nucleate voids at the bond interface, leading to delamination, incomplete bonding, high overlays, and degraded mechanical and electrical integrity. Furthermore, the governing physics of the bonding process involve a strongly coupled, nonlinear fluid–structure interaction that can give rise to non-intuitive behaviors, such as non-monotonic relationships between overlay and initial bond gap. Developing a rigorous predictive model of the bonding process is therefore not merely of academic interest but of direct industrial relevance, as it provides a pathway to understanding and mitigating these failure mechanisms.

The mechanics of wafer bonding were first analyzed using finite element methods by Turner and Spearing [5], who established the plate-bending framework for predicting bonding-induced deformations. The dynamics of the bonding front have been studied since the seminal work of Rieutord et al. [6], who proposed a model relating the bonding front velocity to the surface adhesion energy and the viscous drag of air escaping the shrinking gap between the wafers. Experimental measurements of adhesion energies and bonding wave velocities were subsequently reported by Larrey et al. [7]. This competition between elastic driving forces and viscous resistance was later formalized as a coupled Fluid–Structure Interaction (FSI) problem by Navarro et al. [8], wherein the elastic deformation of the wafer governs the air pressure between the wafers and the resulting air pressure field in turn loads the wafer. The dynamics of the bonding wave has also been analytically examined by Radisson et al. [9], who modeled the propagation speed of the bonding front as a function of the material and geometric parameters. More recently, Jain [10] derived a self-consistent analytical formulation for the bond front velocity by re-parameterizing the viscous dissipation region, providing power-law solutions for the substrate shape during bonding. More recent efforts have employed finite element methods to model various aspects of the bonding process, including bonding wave propagation and cavity-induced deformation modes [11, 12, 13], thermo-mechanical effects in glass-frit bonding [14], and Cu pad expansion in hybrid bonding [15]. Together, these studies have demonstrated the viability of the FSI framework for capturing the essential physics of WxW bonding.

Despite this body of work, existing models share three common limitations: (a) the plate equations employed are not derived from three-dimensional elasticity with explicit statement of the Kirchhoff–Love kinematic hypotheses and their validity criteria; (b) the complete weak formulations necessary for finite element implementation—including the treatment of the fourth-order plate operator, boundary conditions, and discretization details—are not provided, limiting reproducibility; and (c) the inherently nonlinear nature of the coupled system, which gives rise to non-intuitive phenomena such as non-monotonic dependence of bonding wave speed on initial gap, has not been systematically analyzed or attributed to the mathematical structure of the governing equations. Previous formulations generally rely on simplified plate models that are not rigorously derived from the full three-dimensional governing equations of linear elasticity, and they often do not clearly articulate the assumptions under which such dimensional reductions are justified [8, 11]. This gap is significant because the wafer–wafer bonding problem is inherently three-dimensional, involving aspects such as asymmetric incoming wafer geometries, the anisotropic behavior of silicon, and asymmetric boundary conditions driven by how the wafers are constrained and subsequently released. In most cases, the three-dimensional balance of linear momentum, the Kirchhoff–Love kinematic assumption, and the derivation of the flexural rigidity are neither made explicit nor tied to a properly posed boundary value problem. Furthermore, the nonlinear nature of the coupled system has been largely overlooked: the two-way coupling between wafer deflection and air pressure (between the two wafers) yields a genuinely nonlinear, time-dependent FSI problem whose solution behavior is far from intuitive. Non-intuitive phenomena—such as pressure-induced retardation of the bond front and the non-monotone dependence of bonding wave speed on initial bond gap—have not been rigorously analyzed or attributed to the structure of the governing equations.

The present work addresses these shortcomings by developing a mathematically rigorous, fully coupled FSI formulation for WxW bonding, implemented in the high-performance FEniCSx finite element framework [16, 17]. Starting from the three-dimensional equations of linear elasticity for an isotropic wafer, we perform a systematic reduction to the classical Kirchhoff–Love plate equation via through-thickness integration [18], making all assumptions and their validity criteria explicit. The air between the wafers is governed by the Reynolds lubrication equation, and the two-way coupling – wherein the wafer deflection and its rate determine the air pressure, which in turn acts as a distributed transverse load on the wafer – is preserved in full. Notably, the Reynolds pressure field in the bonded region naturally serves as a contact reaction that resists interpenetration of the wafers, eliminating the need for a separate contact algorithm. The resulting inherently nonlinear coupled equations, and the results presented in § 4 demonstrate several counterintuitive implications of this nonlinearity that have not been previously reported or discussed in the WxW bonding literature. The implementation leverages the FEniCSx framework [16, 19, 20, 21], which has proven to be effective for analogous nonlinear multiphysics problems, including nonlinear viscoelasticity [22, 23] and dielectric elastomers [24].

This paper is the first in a planned series. Part I is deliberately restricted to the reduced two-dimensional problem, which suffices for the bonding-front kinetics and air-film pressure studied here, but excludes effects that are intrinsically three-dimensional: in-plane deformation and the resulting post-bond overlay, anisotropic flexural response, asymmetric incoming wafer geometries, and lift-off of the bottom wafer from its chuck, which the present rigid, perfectly adhered bottom wafer precludes by construction. Later parts will progressively address these restrictions, with the eventual objective of solving the full three-dimensional bonding problem. Since the coupling structure established here is unchanged by such extensions, the formulation and monolithic solution strategy of Part I carry over directly; see § 5.

The remainder of this paper is organized as follows. § 2 presents the complete mathematical formulation, beginning with the three-dimensional governing equations and proceeding through their systematic reduction to the two-dimensional coupled plate – Reynolds system. § 3 describes finite element discretization and FEniCSx implementation. § 4 presents numerical results, including validation studies and the non-intuitive physical phenomena arising from the nonlinear coupling. § 5 concludes with a summary and directions for future work.

Theory

The mechanics governing Wafer-to-Wafer (WxW) bonding is characterized by a coupled Fluid-Structure Interaction (FSI) problem. Consider, two wafers in their undeformed initial configurations at time, t=0t = 0, within a Cartesian coordinate system, positioned coaxially by the bonding apparatus on the top and bottom chuck occupying domains Ωtop\Omega_{top} and Ωbot\Omega_{bot} respectively, such that Ωtop={X∈R3:X12+X22≤R2,−tw≤X3≤0}\Omega_{top} = \{\mathbf{X} \in\mathbb{R}^{3} : X_{1}^{2} + X_{2}^{2} \le R^{2}, -t_{w} \le X_{3} \le0\} and Ωbot={X∈R3:X12+X22≤R2,−2tw−h0≤X3≤−tw−h0}\Omega_{bot} = \{\mathbf{X} \in\mathbb{R}^{3} : X_{1}^{2} + X_{2}^{2} \le R^{2}, -2t_{w} - h_{0} \le X_{3} \le-t_{w} - h_{0}\}. The top wafer’s center is located at (0,0,−tw/2)(0,0,-t_{w}/2), and the center of the top surface of the bottom wafer is at (0,0,−3tw/2−h0)(0,0,-3t_{w}/2 - h_{0}), resulting in an initial separation distance h0h_{0} between the bottom surface of the top wafer and the top surface of the bottom wafer. A thin layer of air is entrapped within the interstitial gap between the wafers. At this stage, it is useful to define the following regions, as illustrated in Figure 1

Image not extracted

Figure 1. (a) Cross-sectional view (X1X_1–X3X_3 plane) of the initial undeformed configuration. The top wafer Ωtop\Omega_{top} and bottom wafer Ωbot\Omega_{bot} are separated by an air gap of height h0h_0. Boundary surfaces ∂Ωtopu\partial\Omega^u_{top} and ∂Ωbotl\partial\Omega^l_{bot} are labeled.

(b) Schematic of the wafer-to-wafer bonding process (cross-sectional view): (a) initial configuration with air gap h0h_0 between wafers Ωtop\Omega_{top} and Ωbot\Omega_{bot}; (b) a striker applies force fSf_S to initiate contact at the center; (c) bond front begins to propagate radially outward from the contact zone; (d) bond front continues to advance while entrapped air is expelled; (e) the top wafer is released from the chuck and the striker is retracted; (f) full bonding across the entire interface.

Figure 1: (a) Cross-sectional view of the initial domain configuration showing the top wafer Ωtop\Omega_{top} and the bottom wafer Ωbot\Omega_{bot} separated by an air gap of height h0h_0, and (b) schematic of the six stages of the wafer-to-wafer bonding process.

∂Ωtopu={X:X12+X22≤R2,X3=0},Top surface of the Upper wafer∂Ωtopl={X:X12+X22≤R2,X3=−tw},Bottom surface of the Upper wafer∂Ωtoplat={X:X12+X22=R2,0≤X3≤−tw},Lateral curved surface of the Upper wafer∂Ωbotu={X:X12+X22≤R2,X3=−tw−h0},Top surface of bottom wafer∂Ωbotl={X:X12+X22≤R2,X3=−2tw−h0},Bottom surface of the wafer on the bottom.(1)\begin{aligned} \partial\Omega_{top}^{u} &= \{\mathbf{X} : X_{1}^{2} + X_{2}^{2} \le R^{2}, X_{3} = 0\}, && \text{Top surface of the Upper wafer}\\ \partial\Omega_{top}^{l} &= \{\mathbf{X} : X_{1}^{2} + X_{2}^{2} \le R^{2}, X_{3} = -t_{w}\}, && \text{Bottom surface of the Upper wafer}\\ \partial\Omega_{top}^{lat} &= \{\mathbf{X} : X_{1}^{2} + X_{2}^{2} = R^{2}, 0 \le X_{3} \le-t_{w}\}, && \text{Lateral curved surface of the Upper wafer}\\ \partial\Omega_{bot}^{u} &= \{\mathbf{X} : X_{1}^{2} + X_{2}^{2} \le R^{2}, X_{3} = -t_{w} - h_{0}\}, && \text{Top surface of bottom wafer}\\ \partial\Omega_{bot}^{l} &= \{\mathbf{X} : X_{1}^{2} + X_{2}^{2} \le R^{2}, X_{3} = -2t_{w} - h_{0}\}, && \text{Bottom surface of the wafer on the bottom.} \tag*{(1)} \end{aligned}

Clearly, Ωtop⊂R3\Omega_{top} \subset\mathbb{R}^{3} denotes the three-dimensional reference domain occupied by the top wafer. For the sake of completeness for discussions in the following sections, we introduce

ω={xˉ=(X1,X2)∈R2:X12+X22≤R2}(2)\omega= \{\bar{\mathbf{x}} = (X_{1}, X_{2}) \in\mathbb{R}^{2}: X_{1}^{2} + X_{2}^{2} \le R^{2}\} \tag*{(2)}

as the two-dimensional mid-surface of the top wafer, obtained by projecting Ωtop\Omega_{top} onto the plane X3=0X_{3} = 0.2

The subsequent bonding process encompasses the following four stages:

  • Initiation: The process commences when a striker of radius RSR_{S} applies a localized force, fS(X,t)=fS(X,t)e3\mathbf{f}_{S}(\mathbf{X},t)=f_{S}(\mathbf{X},t)\mathbf{e}_{3} at time t=0t=0, where X∈∂ΩS={X∈∂Ωtop:X12+X22≤RS2,X3=0}\mathbf{X}\in\partial\Omega_{S}=\{\mathbf{X}\in\partial\Omega_{top}:X_{1}^{2}+X_{2}^{2}\le R_{S}^{2},X_{3}=0\}, to the central region of the top wafer. This applied force induces a deformation in the top wafer, causing it to make initial physical contact with the bottom wafer at the point of maximum deflection and thereby modifying the effective separation h(X,t)h(\mathbf{X},t) between the wafers from its initial value h(X,t=0)=h0h(\mathbf{X},t=0)=h_{0}.

  • Bond Propagation: Following initiation, the bond front propagates radially outward. The interfacing surfaces of both wafers undergo a pretreatment process (e.g., plasma activation) to establish attractive interfacial forces fB(X,h(X,t))\mathbf{f}_{B}(\mathbf{X},h(\mathbf{X},t)). These forces become significant when the local separation between the wafer surfaces reduces below a critical threshold δ\delta such that

fB(X,h(X,t))=fB(X,h)={fBE,if h(X,t)≤δ,0,otherwise.(3)\mathbf{f}_{B}(\mathbf{X},h(\mathbf{X},t))=\mathbf{f}_{B}(\mathbf{X},h)= \begin{cases} \mathbf{f}_{BE}, & \text{if } h(\mathbf{X},t)\le\delta,\\ 0, & \text{otherwise}. \end{cases} \tag*{(3)}

Consequently, these attractive forces draw the top wafer progressively towards the bottom wafer, concurrently expelling the entrapped air from the advancing bond interface. This constitutive assumption is adopted throughout the remainder of the paper.

  • Wafer Release: Subsequently, the top wafer is released from the retaining chuck of the bonding apparatus. This release allows the remainder of the top wafer to settle onto the bottom wafer under the influence of the propagating interfacial forces and potentially gravitational effects, facilitating the continued expansion of the bonded area.

  • Bond Completion: The process culminates when the bond interface has propagated across the entire surface of the wafers, resulting in a fully bonded wafer pair.

The three-dimensional governing equations

Under the assumption of a rigid bottom wafer, the deformation of the top wafer is described by the displacement field

u(X,t)=(u1(X,t),u2(X,t),w(X,t)),X∈Ωtop, t∈(0,T],(4)\mathbf{u}(\mathbf{X},t)=(u_{1}(\mathbf{X},t),u_{2}(\mathbf{X},t),w(\mathbf{X},t)), \qquad\mathbf{X}\in\Omega_{top},\ t\in(0,T], \tag*{(4)}

where u1u_{1}, u2u_{2} denote the in-plane components and ww is the out-of-plane (transverse) component. The linearized strain tensor associated with (4) is

ε(u)=12(∇u+∇uT).(5)\boldsymbol{\varepsilon}(\mathbf{u})=\frac{1}{2}\left(\nabla\mathbf{u}+\nabla\mathbf{u}^{\mathsf{T}}\right). \tag*{(5)}

For an isotropic, linearly elastic material, the Cauchy stress tensor is given by the constitutive relation

σ=σ(ε(u))=2Gε(u)+λtr⁡(ε(u))I,(6)\boldsymbol{\sigma}=\boldsymbol{\sigma}(\boldsymbol{\varepsilon}(\mathbf{u}))=2G\boldsymbol{\varepsilon}(\mathbf{u})+\lambda\operatorname{tr}(\boldsymbol{\varepsilon}(\mathbf{u}))\mathbf{I}, \tag*{(6)}

where G=E/[2(1+ν)]G=E/[2(1+\nu)] and λ=Eν/[(1+ν)(1−2ν)]\lambda=E\nu/[(1+\nu)(1-2\nu)] are the Lamé parameters, EE is Young’s modulus, and ν\nu is Poisson’s ratio.

Under quasi-static conditions, the balance of linear momentum for the top wafer, incorporating all body and surface forces, reads

div⁡σ(X,t)=b(X,t)∀(X,t)∈Ωtop×(0,T],(7)\operatorname{div}\boldsymbol{\sigma}(\mathbf{X},t)=\mathbf{b}(\mathbf{X},t)\quad\forall(\mathbf{X},t)\in\Omega_{\mathrm{top}}\times(0,T], \tag*{(7)}
σn={0,∀(X,t)∈∂Ωtoplat×(0,T],fS(X,t)e3,∀(X,t)∈∂ΩS×(0,T],fC(X,t),∀(X,t)∈∂Ωtopu×(0,T],fC(X,t)+fB(X,t)+p(X,t)n(x,t),∀(X,t)∈∂Ωtopl×(0,T],(8)\boldsymbol{\sigma}\mathbf{n}= \begin{cases} \mathbf{0}, & \forall(\mathbf{X},t)\in\partial\Omega_{\mathrm{top}}^{\mathrm{lat}}\times(0,T],\\ f_{\mathrm{S}}(\mathbf{X},t)\mathbf{e}_3, & \forall(\mathbf{X},t)\in\partial\Omega_{\mathrm{S}}\times(0,T],\\ \mathbf{f}^{\mathrm{C}}(\mathbf{X},t), & \forall(\mathbf{X},t)\in\partial\Omega_{\mathrm{top}}^{\mathrm{u}}\times(0,T],\\ \mathbf{f}_{\mathrm{C}}(\mathbf{X},t)+\mathbf{f}_{\mathrm{B}}(\mathbf{X},t)+p(\mathbf{X},t)\mathbf{n}(\mathbf{x},t), & \forall(\mathbf{X},t)\in\partial\Omega_{\mathrm{top}}^{\mathrm{l}}\times(0,T], \end{cases} \tag*{(8)}

where:

  • b(X,t)\mathbf{b}(\mathbf{X},t) is the body-force density (e.g., b={0,0,b3}=−ρge3\mathbf{b}=\{0,0,b_3\}=-\rho g\mathbf{e}_3),

  • fC(∂Ωtopu,t)\mathbf{f}^{\mathrm{C}}(\partial\Omega_{\mathrm{top}}^{\mathrm{u}},t) is the contact traction exerted by the chuck on the top surface of the bonding apparatus,

  • fB(X,h)\mathbf{f}_{\mathrm{B}}(\mathbf{X},h) is the bonding traction (cf. (3)),

  • fC(X,t)\mathbf{f}_{\mathrm{C}}(\mathbf{X},t) is the contact traction preventing inter-penetration of the wafers,

  • p(X,t)p(\mathbf{X},t) is the pressure of the air in the bond gap, which induces the normal load p(X,t)n(X,t)p(\mathbf{X},t)\mathbf{n}(\mathbf{X},t) on ∂Ωtopl\partial\Omega_{\mathrm{top}}^{\mathrm{l}}.

The air pressure pp is determined by the three-dimensional Reynolds lubrication equation posed on the gap domain Ωgap\Omega_{\mathrm{gap}} with gap height h(X,t)=h0+w(X,t)h(\mathbf{X},t)=h_0+w(\mathbf{X},t):

12μ∂h∂t−∇⋅(h3∇p)=0,∀X∈Ωgap, t∈(0,T],(9)12\mu\frac{\partial h}{\partial t}-\nabla\cdot\left(h^3\nabla p\right)=0,\quad\forall\mathbf{X}\in\Omega_{\mathrm{gap}},\ t\in(0,T], \tag*{(9)}

where μ\mu is the dynamic viscosity of air.

The use of the Reynolds lubrication equation is justified by the following conditions, all of which are satisfied in the wafer bonding configuration [25]:

  1. Thin-film assumption: The gap height is much smaller than the lateral extent, ε=h0/R∼100 μm/150 mm≈6.7×10−4≪1\varepsilon=h_0/R\sim100\,\mu\mathrm{m}/150\,\mathrm{mm}\approx6.7\times10^{-4}\ll1 (assuming a large gap of 100 μm100\,\mu\mathrm{m}).

  2. Low Reynolds number: Viscous forces dominate over inertial forces. Taking the bond-front velocity V∼10 mm/sV\sim10\,\mathrm{mm}/\mathrm{s} as the characteristic velocity, Re⁡=ρairVh0/μ≈(1.2)(0.01)(10−4)/(1.81×10−5)≈0.07≪1\operatorname{Re}=\rho_{\mathrm{air}}Vh_0/\mu\approx(1.2)(0.01)(10^{-4})/(1.81\times10^{-5})\approx0.07\ll1.

  3. Quasi-steady pressure: The acoustic time scale for pressure equilibration, τacoustic∼h0/csound≈0.3 μs\tau_{\mathrm{acoustic}}\sim h_0/c_{\mathrm{sound}}\approx0.3\,\mu\mathrm{s}, is orders of magnitude smaller than the mechanical time scale (∼1\sim1–10 s10\,\mathrm{s}).

  4. Continuum validity: The Knudsen number Kn=λ/h\mathrm{Kn}=\lambda/h, where λ\lambda denotes the molecular mean free path of the gas, classifies the flow regime: Kn<0.01\mathrm{Kn}<0.01 corresponds to continuum flow, 0.01<Kn<0.10.01<\mathrm{Kn}<0.1 to the slip-flow regime, and Kn>0.1\mathrm{Kn}>0.1 to the transitional regime where the continuum-based Reynolds equation breaks down [26, 27]. For air at atmospheric pressure and room temperature, λ≈68 nm\lambda\approx68\,\mathrm{nm}. At the initial gap h0=100 μmh_0=100\,\mu\mathrm{m}, Kn≈6.8×10−4≪0.01\mathrm{Kn}\approx6.8\times10^{-4}\ll0.01, placing the flow well within the continuum regime. As the gap closes during bonding, the local

Knudsen number increases; however, the simultaneously rising air pressure near the bond front reduces λ\lambda (since λ∝1/P\lambda\propto1/P from kinetic theory) [28], partially mitigating the increase in Kn. The regularization introduced in § 3 ensures that the effective gap h^\hat{h} remains sufficiently large to prevent the flow from entering the transitional regime.

The coupled three-dimensional FSI problem can thus be stated abstractly as:

G1(u,p)≡div⁡σ(u)−b=0,(10)\mathcal{G}_1(\mathbf{u},p) \equiv\operatorname{div}\boldsymbol{\sigma}(\mathbf{u})-\mathbf{b}=0, \tag*{(10)}
G2(p,u)≡12μ∂h∂t−∇⋅(h3∇p)=0,(11)\mathcal{G}_2(p,\mathbf{u}) \equiv12\mu\frac{\partial h}{\partial t}-\nabla\cdot\left(h^3\nabla p\right)=0, \tag*{(11)}

subjected to the traction boundary conditions in Eq. (7b). Here, the coupling is two-way: the displacement u\mathbf{u} determines the gap hh appearing in G2\mathcal{G}_2, and the pressure pp enters as a surface traction in G1\mathcal{G}_1. This system of governing equations shares structural similarities with models for the nonlinear viscoelasticity of rubber-like solids [22, 23] and the transient response of dielectric elastomers [24]. Such problems have been successfully implemented using the legacy FEniCS computing platform and, more recently, the high-performance FEniCSx framework [17].

Reduction to the Kirchhoff–Love plate model

Following [18], the three-dimensional problem G1=0\mathcal{G}_1=0, G2=0\mathcal{G}_2=0 and the traction boundary conditions in Eq. (7b) can be reduced to a two-dimensional plate model in the thin-wafer limit tw/R≪1t_w/R \ll1. The classical Kirchhoff–Love hypothesis asserts:

  1. Straight material fibers initially normal to the mid-surface remain straight and normal to the deformed mid-surface (i.e., the transverse shear strains vanish, εα3=0\varepsilon_{\alpha3}=0, α=1,2\alpha=1,2).

  2. The thickness of the plate does not change during deformation (i.e., ε33=0\varepsilon_{33}=0).

Kinematics. Under these hypotheses, the three-dimensional displacement field (4) reduces to the form

uα(X,t)=−X3∂w∂Xα(xˉ,t),α=1,2,u3(X,t)=w(xˉ,t),(12)u_\alpha(\mathbf{X},t)=-X_3\frac{\partial w}{\partial X_\alpha}(\bar{\mathbf{x}},t),\quad\alpha=1,2,\qquad u_3(\mathbf{X},t)=w(\bar{\mathbf{x}},t), \tag*{(12)}

where xˉ=(X1,X2)\bar{\mathbf{x}}=(X_1,X_2) are the in-plane coordinates of the mid-surface ω\omega (cf. (2)) and w(xˉ,t)w(\bar{\mathbf{x}},t) is the transverse deflection of the mid-surface. In particular, the in-plane displacements are entirely determined by ww and its gradient, so that at the mid-surface (X3=0X_3=0)

u⟶(0,0,w(xˉ,t)).(13)\mathbf{u}\longrightarrow(0,0,w(\bar{\mathbf{x}},t)). \tag*{(13)}

Substituting (12) into the strain definition (5), the only non-vanishing components of ε\boldsymbol{\varepsilon} are

εαβ=−X3καβ(xˉ,t),καβ=∂2w∂Xα∂Xβ,α,β=1,2,(14)\varepsilon_{\alpha\beta}=-X_3\kappa_{\alpha\beta}(\bar{\mathbf{x}},t),\qquad\kappa_{\alpha\beta}=\frac{\partial^2w}{\partial X_\alpha\partial X_\beta},\quad\alpha,\beta=1,2, \tag*{(14)}

where κ=∇2w\boldsymbol{\kappa}=\nabla^2w is the curvature tensor of the mid-surface.

Bending moments. Invoking the plane-stress condition (σ33=0\sigma_{33}=0) appropriate for thin plates, the in-plane constitutive relation reads

σαβ=E1−ν2[(1−ν)εαβ+νεγγδαβ],α,β=1,2.(15)\sigma_{\alpha\beta}=\frac{E}{1-\nu^2}\left[(1-\nu)\varepsilon_{\alpha\beta}+\nu\varepsilon_{\gamma\gamma}\delta_{\alpha\beta}\right],\quad\alpha,\beta=1,2. \tag*{(15)}

The bending-moment tensor is obtained by integrating the first moment of the in-plane stress through the thickness:

Mαβ=∫−tw/2tw/2X3σαβ dX3=D[(1−ν)καβ+νκγγδαβ],(16)M_{\alpha\beta}=\int_{-t_w/2}^{t_w/2}X_3\sigma_{\alpha\beta}\,\mathrm{d}X_3=D\left[(1-\nu)\kappa_{\alpha\beta}+\nu\kappa_{\gamma\gamma}\delta_{\alpha\beta}\right], \tag*{(16)}
SymbolMeaning
Geometric/kinematic parameters
X\mathbf{X}Reference (material) coordinate in R3\mathbb{R}^{3}; tt denotes time.
xˉ\bar{\mathbf{x}}In-plane coordinate on the mid-surface ω\omega.
Ωtop\Omega_{top}Three-dimensional reference domain of the top wafer.
Ωbot\Omega_{bot}Three-dimensional reference domain of the bottom wafer.
Ωgap\Omega_{gap}Thin interstitial air-gap domain between the wafers.
ω\omegaTwo-dimensional mid-surface domain of the top wafer (projection of Ωtop\Omega_{top} onto X3=0X_{3}=0).
∂Ωtopu,l\partial\Omega_{top}^{u,l}Top/bottom surfaces of the top wafer; ∂Ωtoplat\partial\Omega_{top}^{lat} its lateral surface.
ΓR\Gamma_{R}Outer edge of the plate/mid-surface, ΓR={xˉ∈R2:X12+X22=R2}\Gamma_{R}=\{\bar{\mathbf{x}}\in\mathbb{R}^{2}:X_{1}^{2}+X_{2}^{2}=R^{2}\}.
Γvent\Gamma_{\mathrm{vent}}Vented portion of the boundary where the air gap is open to ambient; here Γvent=ΓR\Gamma_{\mathrm{vent}}=\Gamma_{R}.
n\mathbf{n}Outward unit normal.
RRWafer radius.
twt_{w}Wafer thickness.
h0h_{0}Initial air-gap height (initial wafer separation).
h(X,t)h(\mathbf{X},t)Local gap between wafers (here h=h0+wh=h_{0}+w).
rB(t)r_{B}(t)Bond-front radius, i.e. the radial position of the moving bonded/unbonded interface.
RS/RpinR_{S}/R_{pin}Striker/pin radius.
TTFinal time of interest for t∈(0,T]t\in(0,T].
u(X,t)\mathbf{u}(\mathbf{X},t)Displacement field; ww denotes the transverse displacement component.
ε(u)\boldsymbol{\varepsilon}(\mathbf{u})Linearized strain tensor.
σ\boldsymbol{\sigma}Cauchy stress tensor.
Constitutive parameters
E,νE,\nuYoung’s modulus and Poisson’s ratio.
G,λG,\lambdaLamé parameters.
DDPlate flexural rigidity, D=Etw3/[12(1−ν2)]D=Et_{w}^{3}/[12(1-\nu^{2})].
ρ\rhoMass density.
μ\muDynamic viscosity of air.
Force/loading parameters
b(X,t)\mathbf{b}(\mathbf{X},t)Body-force density (e.g., gravity, b=−ρge3\mathbf{b}=-\rho g\mathbf{e}_{3}).
ggGravitational acceleration.
fC(X,t)\mathbf{f}^{C}(\mathbf{X},t)Chuck contact traction applied on ∂Ωtopu\partial\Omega_{top}^{u}.
fC(X,t)\mathbf{f}_{C}(\mathbf{X},t)Unilateral contact traction preventing inter-penetration.
fS(X,t)=fS(X,t)e3\mathbf{f}_{S}(\mathbf{X},t)=f_{S}(\mathbf{X},t)\mathbf{e}_{3}Striker-applied surface traction/force used to initiate contact.
qext(xˉ,t)q_{\mathrm{ext}}(\bar{\mathbf{x}},t)Resultant transverse external loading per unit area of the mid-surface.
δ\deltaCritical separation below which bonding forces activate.
fB(X,h)\mathbf{f}_{B}(\mathbf{X},h)Interfacial bonding traction as a function of gap.
fBEf_{BE}Constant bonding traction for h≤δh\le\delta.
Γ=fBEδ\Gamma=f_{BE}\deltaConstant bonding energy h≤δh\le\delta.
p(X,t)p(\mathbf{X},t)Pressure of the entrapped air in the gap.
patmp_{\mathrm{atm}}Ambient (atmospheric) pressure prescribed on Γvent\Gamma_{\mathrm{vent}}.

Table 1. Parameters appearing in § 2.

where

D=Etw312(1−ν2)(17)D = \frac{E t_w^3}{12(1-\nu^2)} \tag*{(17)}

is the flexural rigidity of the plate.

Equilibrium. Integration of the three-dimensional balance of linear momentum (7) through the thickness, together with the traction boundary conditions on ∂Ωtopu\partial\Omega^u_{top} and ∂Ωtopl\partial\Omega^l_{top}, yields the classical Kirchhoff–Love plate equation (see, e.g., [18] and references therein):

DΔ2w−qext+p−fS−fC(w)−fB(w)−fC(w)=0∀xˉ∈ω, t∈(0,T],(18)D \Delta^2 w - q_{\mathrm{ext}} + p - f_S - f^C(w) - f_B(w) - f_C(w) = 0 \qquad\forall\bar{x} \in\omega,\ t \in(0,T], \tag*{(18)}

where Δ2≡Δ(Δ)\Delta^2 \equiv\Delta(\Delta) is the biharmonic operator and:

  • qext(xˉ,t)q_{\mathrm{ext}}(\bar{x},t) is the resultant transverse external loading per unit area of the mid-surface, obtained by through-thickness integration of the body force:

qext(xˉ,t)=∫−tw/2tw/2b3 dX3,(19)q_{\mathrm{ext}}(\bar{x},t) = \int_{-t_w/2}^{t_w/2} b_3\,\mathrm{d}X_3, \tag*{(19)}

(Note b3:=b⋅e3=ρgb_3 := \mathbf{b}\cdot\mathbf{e}_3 = \rho g denotes the transverse (X3X_3) component of the body-force density b\mathbf{b}.)

  • p(xˉ,t)p(\bar{x},t) is the air pressure in the gap,

  • fB(w)f_B(w) is the bonding traction per unit area, and

  • fC(w)f^C(w) and fC(w)f_C(w) is the unilateral contact forces per unit area as explained in § 2.5.

Equation (18) is a fourth-order PDE in ww that inherits, through the through-thickness integration, all body and surface forces from the original three-dimensional governing equation (7). Note that, in these expressions, the boldface notation for the forces in (7) has been omitted to emphasize that the corresponding terms in the plate equation (18) are scalar transverse components of the forces.

In the same thin-wafer limit, the Reynolds equation (9) reduces to a two-dimensional equation on the mid-surface ω\omega. Introducing the gap

h(xˉ,t)=h0+w(xˉ,t),(20)h(\bar{x},t) = h_0 + w(\bar{x},t), \tag*{(20)}

with h0h_0 denoting the initial uniform separation, the two-dimensional pressure equation is

12μ∂h∂t−∇⋅(h3∇p)=0∀xˉ∈ω, t∈(0,T].(21)12\mu\frac{\partial h}{\partial t} - \nabla\cdot\left(h^3\nabla p\right) = 0 \qquad\forall\bar{x} \in\omega,\ t \in(0,T]. \tag*{(21)}

The reduced two-dimensional coupled FSI problem is therefore stated abstractly as:

G~1(w(xˉ,t),p(xˉ,t))≡DΔ2w−qext+p−fS−fC(w)−fB(w)−fC(w)=0∀xˉ∈ω, t∈(0,T],(22)\widetilde{\mathcal{G}}_1(w(\bar{x},t),p(\bar{x},t)) \equiv D\Delta^2w-q_{\mathrm{ext}}+p-f_S-f^C(w)-f_B(w)-f_C(w)=0 \qquad\forall\bar{x}\in\omega,\ t\in(0,T], \tag*{(22)}
G~2(p(xˉ,t),w(xˉ,t))≡12μ∂(h0+w)∂t−∇⋅((h0+w)3∇p)=0∀xˉ∈ω, t∈(0,T].(23)\widetilde{\mathcal{G}}_2(p(\bar{x},t),w(\bar{x},t)) \equiv12\mu\frac{\partial(h_0+w)}{\partial t}-\nabla\cdot\left((h_0+w)^3\nabla p\right)=0 \qquad\forall\bar{x}\in\omega,\ t\in(0,T]. \tag*{(23)}

Boundary and initial conditions

The coupled system (22)–(23) is supplemented with the following boundary and initial conditions for uniqueness of solutions. The boundary conditions described below apply to the top wafer, which is the only deformable body in the model. The bottom wafer is treated as rigid and fixed to the bottom chuck; its upper surface defines the reference plane X3=−tw−h0X_3=-t_w-h_0 against which contact is enforced.

Mechanical boundary conditions for the plate. On the outer edge ΓR={xˉ∈R2:X12+X22=R2}\Gamma_R=\{\bar{x}\in\mathbb{R}^2:X_1^2+X_2^2=R^2\}:

w(xˉ,t)=wΓ(xˉ,t)∀xˉ∈ΓR, t∈(0,T],(24)w(\bar{x},t)=w_{\Gamma}(\bar{x},t) \qquad\forall\bar{x}\in\Gamma_R,\ t\in(0,T], \tag*{(24)}
∂w∂n(xˉ,t)=0∀xˉ∈ΓR, t∈(0,T],(25)\frac{\partial w}{\partial n}(\bar{x},t)=0 \qquad\forall\bar{x}\in\Gamma_R,\ t\in(0,T], \tag*{(25)}

where wΓw_{\Gamma} is the prescribed displacement at the wafer edge (e.g., from the top chuck constraint or the release of the top wafer from the top chuck) and a zero-slope condition is imposed for a clamped boundary.

On any symmetry boundaries Γs\Gamma_s (exploited in the quarter-wafer model):

∂w∂n(xˉ,t)=0,Mnn(w)=0∀xˉ∈Γs, t∈(0,T].(26)\frac{\partial w}{\partial n}(\bar{x},t)=0,\qquad M_{nn}(w)=0 \qquad\forall\bar{x}\in\Gamma_s,\ t\in(0,T]. \tag*{(26)}

Pressure boundary and initial conditions. Let Γvent⊆∂ω\Gamma_{\mathrm{vent}}\subseteq\partial\omega denote the portion of the mid-surface boundary where the air gap between the wafers is connected to the ambient environment (i.e., the air pressure is prescribed). In the fully vented configuration considered here, we take Γvent=ΓR\Gamma_{\mathrm{vent}}=\Gamma_R.

p(xˉ,t)=patm∀xˉ∈Γvent, t∈(0,T],(27)p(\bar{x},t)=p_{\mathrm{atm}} \qquad\forall\bar{x}\in\Gamma_{\mathrm{vent}},\ t\in(0,T], \tag*{(27)}
p(xˉ,0)=p0(xˉ)∀xˉ∈ω,(28)p(\bar{x},0)=p_0(\bar{x}) \qquad\forall\bar{x}\in\omega, \tag*{(28)}

where patmp_{\mathrm{atm}} is the ambient (atmospheric) pressure and p0p_0 is the initial pressure distribution (typically p0=patmp_0=p_{\mathrm{atm}}).

Initial condition for the deflection:

w(xˉ,0)=0∀xˉ∈ω.(29)w(\bar{x},0)=0 \qquad\forall\bar{x}\in\omega. \tag*{(29)}

Summary of the coupled problem

For convenience, the complete coupled plate–air FSI problem is collected here. Find the transverse deflection w(xˉ,t)w(\bar{x},t) and air pressure p(xˉ,t)p(\bar{x},t) on the mid-surface domain ω\omega for t∈(0,T]t\in(0,T] satisfying:

G~1(w(xˉ,t),p(xˉ,t))≡DΔ2w−qext+p−fS−fC(w)−fB(w)−fC(w)=0∀xˉ∈ω, t∈(0,T],(30)\widetilde{\mathcal{G}}_1(w(\bar{x},t),p(\bar{x},t))\equiv D\Delta^2w-q_{\mathrm{ext}}+p-f_S-f^C(w)-f^B(w)-f^C(w)=0 \qquad\forall\bar{x}\in\omega,\ t\in(0,T], \tag*{(30)}
G~2(p(xˉ,t),w(xˉ,t))≡12μ∂h∂t−∇⋅(h3∇p)=0∀xˉ∈ω, t∈(0,T],(31)\widetilde{\mathcal{G}}_2(p(\bar{x},t),w(\bar{x},t))\equiv12\mu\frac{\partial h}{\partial t}-\nabla\cdot(h^3\nabla p)=0 \qquad\forall\bar{x}\in\omega,\ t\in(0,T], \tag*{(31)}
h(xˉ,t)=h0+w(xˉ,t)∀xˉ∈ω, t∈(0,T],(32)h(\bar{x},t)=h_0+w(\bar{x},t) \qquad\forall\bar{x}\in\omega,\ t\in(0,T], \tag*{(32)}

with the boundary conditions (21)–(26) for ww, the pressure conditions (23), and the initial conditions (29) and (28). The coupling is two-way: the deflection ww modifies the gap hh in G~2\widetilde{\mathcal{G}}_2 and thereby the air-flow resistance, while the air pressure pp enters the plate equation G~1\widetilde{\mathcal{G}}_1 as a distributed transverse load.

The Contacts

We model contact with the rigid chucks/wafers via (penalty-type) tractions expressed in terms of the transverse deflection ww.

Top-chuck contact. Let ktopC>0k_{\mathrm{top}}^{C} > 0 denote a penalty stiffness associated with the rigid top chuck. The corresponding contact traction fCf^{C} is defined by

fC(xˉ,t)≡fC(w(xˉ,t))={ktopCw(xˉ,t),if w(xˉ,t)>0,0,otherwise.(33)f^{C}(\bar{x},t) \equiv f^{C}(w(\bar{x},t)) = \begin{cases} k_{\mathrm{top}}^{C} w(\bar{x},t), & \text{if } w(\bar{x},t) > 0, \\ 0, & \text{otherwise.} \end{cases} \tag*{(33)}

This traction penalizes (and thereby prevents) deformation states with w(xˉ,t)>0w(\bar{x},t) > 0 that would violate the geometric constraint imposed by the rigid top chuck.

Bottom-wafer contact. We also introduce a unilateral contact traction fCf_{C} to prevent interpenetration of the top wafer into the bottom wafer. In general, fCf_{C} may be defined by an analogous penalty law activated at the closure of the gap. However, as shown in § 4.2, the Reynold’s pressure pp is sufficient to enforce the non-penetration constraint in the bonded regions and therefore provides an effective contact penalty. Accordingly, throughout the remainder of the paper we set

fC(w)=0.(34)f_{C}(w) = 0. \tag*{(34)}

Bottom wafer–chuck contact. Contact between the bottom wafer and the bottom chuck is implicitly enforced through the rigid-wafer assumption: the bottom wafer remains stationary throughout the simulation, representing perfect adhesion to the chuck.

FEniCSx Implementation

This section details the finite element discretization and solution of the coupled plate–air system (25) using FEniCSx [16]. Exploiting the four-fold symmetry of the circular wafer, we solve on a quarter-domain ωq={xˉ∈ω:X1≥0,X2≥0}\omega_q = \{\bar{x} \in\omega: X_1 \ge0, X_2 \ge0\} with appropriate symmetry conditions. The plate equation G~1=0\widetilde{G}_1 = 0 is discretized with a C0C^0 interior-penalty (C0IP) Kirchhoff–Love formulation [29], and the Reynolds equation G~2=0\widetilde{G}_2 = 0 is discretized with standard continuous Galerkin finite elements. Both fields are solved monolithically.

Recap of the strong-form equations

For convenience we restate the coupled system (25). Find w(xˉ,t)w(\bar{x},t) and p(xˉ,t)p(\bar{x},t) satisfying:

Plate equation (from (30)):

DΔ2w(xˉ,t)−qext(xˉ,t)+p(xˉ,t)−fS−fC(w)−fB(w)−fC(w)=0∀xˉ∈ωq, t∈(0,T],(35)D \Delta^2 w(\bar{x},t) - q_{\mathrm{ext}}(\bar{x},t) + p(\bar{x},t) - f_{S} - f^{C}(w) - f_{B}(w) - f_{C}(w) = 0 \qquad\forall\bar{x} \in\omega_q,\ t \in(0,T], \tag*{(35)}

where DD is the flexural rigidity (17) and the bending-moment tensor MαβM_{\alpha\beta} is given by (16). In operator form, DΔ2w=Mαβ,αβD \Delta^2 w = M_{\alpha\beta,\alpha\beta}, i.e.,

M(w)=D[(1−ν)κ(w)+νtr⁡(κ(w))I],κ(w)=∇2w.(36)\mathbf{M}(w) = D[(1-\nu)\boldsymbol{\kappa}(w) + \nu\operatorname{tr}(\boldsymbol{\kappa}(w))\mathbf{I}], \qquad\boldsymbol{\kappa}(w) = \nabla^2 w. \tag*{(36)}

Reynolds equation (from (31)–(32)):

12μ∂h∂t(xˉ,t)−∇⋅(h^(xˉ,t)3∇p(xˉ,t))=0∀xˉ∈ωq, t∈(0,T],(37)12\mu\frac{\partial h}{\partial t}(\bar{x},t) - \nabla\cdot\left(\hat{h}(\bar{x},t)^3 \nabla p(\bar{x},t)\right) = 0 \qquad\forall\bar{x} \in\omega_q,\ t \in(0,T], \tag*{(37)}

with h^(xˉ,t)=h0+(1−β)w(xˉ,t)\hat{h}(\bar{x},t) = h_0 + (1-\beta)w(\bar{x},t). Here, 0<β≤0.00750 < \beta\le0.0075 is a small dimensionless regularization parameter; this choice is necessary to prevent pp from becoming unphysical as the gap closes, w→−h0w \to-h_0.

The regularization parameter β\beta maintains a minimum effective gap h^min⁡=βh0\hat{h}_{\min} = \beta h_0 when full contact occurs (w→−h0w \to-h_0). This regularization has both physical and numerical justifications:

  • Physical basis: Even for highly polished silicon wafers, sub-micron surface roughness prevents perfect molecular contact. The root-mean-square roughness of CMP-polished wafers is typically 0.1–0.5 nm [30], but asperity peaks can extend to several nanometers. The regularized minimum gap (βh0≈0.75 μm\beta h_0 \approx0.75\,\mu\mathrm{m} for h0=100 μmh_0 = 100\,\mu\mathrm{m} and β=0.0075\beta= 0.0075) is conservative relative to this physical roughness scale.

  • Numerical justification: The h3h^3 term in the Reynolds equation causes numerical ill-conditioning as h→0h \to0. The regularization ensures that the discrete system remains well-posed and that the Newton solver converges reliably.

  • Continuum validity: As discussed in § 2 (condition (iv)), the Knudsen number Kn=λ/h\mathrm{Kn} = \lambda/h governs the transition between continuum and rarefied gas flow regimes [26, 27]. At the regularized minimum gap h^min⁡=βh0=0.75 μm\hat{h}_{\min} = \beta h_0 = 0.75\,\mu\mathrm{m} (β=0.0075\beta= 0.0075, h0=100 μmh_0 = 100\,\mu\mathrm{m}), the Knudsen number at atmospheric pressure is Kn≈68 nm/0.75 μm≈0.09\mathrm{Kn} \approx68\,\mathrm{nm}/0.75\,\mu\mathrm{m} \approx0.09; however, the simultaneously rising pressure near the bond front reduces λ\lambda (since λ∝1/P\lambda\propto1/P), keeping the flow within the slip-flow regime (Kn<0.1\mathrm{Kn} < 0.1) where the Reynolds equation remains applicable [26].

  • Asymptotic validity: The mathematical validity of the Reynolds equation as a rigorous limit of the Navier–Stokes equations is established through asymptotic analysis under small aspect ratios and geometric perturbations [31]. Perturbing the film thickness by the small scaling factor β=0.0075\beta= 0.0075 confines the introduced error to O(β2)O(\beta^2), well within the accepted tolerances of the thin-film approximation.

A sensitivity analysis demonstrating that the results are insensitive to β\beta in the range [0.0025, 0.01] is presented in Appendix B.

Boundary conditions. On ωq\omega_q the boundary ∂ωq\partial\omega_q is decomposed into:

  • Symmetry edges Γx={xˉ∈∂ωq:X1=0}\Gamma_x = \{\bar{x} \in\partial\omega_q : X_1 = 0\} and Γy={xˉ∈∂ωq:X2=0}\Gamma_y = \{\bar{x} \in\partial\omega_q : X_2 = 0\}, where the slope vanishes, ∂nw=0\partial_n w = 0, and Mnn(w)=0M_{nn}(w) = 0 (cf. (26)).

  • Outer arc ΓR={xˉ∈∂ωq:X12+X22=R2}\Gamma_R = \{\bar{x} \in\partial\omega_q : X_1^2 + X_2^2 = R^2\}, where displacement Dirichlet conditions w=wΓw = w_\Gamma (cf. (24)) and the pressure venting condition pair=p∞p_{\mathrm{air}} = p_\infty (cf. (27)) are imposed.

Time discretization

The Reynolds equation (37) is the only equation containing a time derivative. We discretize it with an implicit (backward) Euler scheme. Partitioning (0,T](0,T] into time steps 0=t0<t1<⋯<tN=T0 = t_0 < t_1 < \cdots< t_N = T with Δtn=tn+1−tn\Delta t_n = t_{n+1} - t_n, and denoting wn≈w(⋅,tn)w^n \approx w(\cdot,t_n), pn+1≈pair(⋅,tn+1)p^{n+1} \approx p_{\mathrm{air}}(\cdot,t_{n+1}), the semi-discrete system at step n+1n+1 reads:

DΔ2wn+1−qextn+1+pairn+1−fS(wn+1)−fC(wn+1)−fB(wn+1)−fC(wn+1)=0∀ xˉ∈ωq,(38)D \Delta^2 w^{n+1} - q_{\mathrm{ext}}^{n+1} + p_{\mathrm{air}}^{n+1} - f_{\mathrm{S}}(w^{n+1}) - f^{\mathrm{C}}(w^{n+1}) - f_{\mathrm{B}}(w^{n+1}) - f_{\mathrm{C}}(w^{n+1}) = 0 \quad\forall\,\bar{x} \in\omega_q, \tag*{(38)}
12μΔtn(wn+1−wn)−∇⋅(h^(wn+1)3∇pn+1)=0∀ xˉ∈ωq.(39)\frac{12\mu}{\Delta t_n}(w^{n+1} - w^n) - \nabla\cdot\left(\hat{h}(w^{n+1})^3 \nabla p^{n+1}\right) = 0 \quad\forall\,\bar{x} \in\omega_q. \tag*{(39)}

Weak formulation

Let V=H2(ωq)V = H^2(\omega_q) and Q=H1(ωq)Q = H^1(\omega_q) denote the trial/test spaces for the deflection and pressure fields, respectively, and let v∈Vv \in V, q∈Qq \in Q be test functions. Multiplying (38) by vv, integrating by parts twice, and multiplying (39) by qq with a single integration by parts, yields the following weak residuals.

Image not extracted

Figure 2. Quarter-domain ωq\omega_q exploiting four-fold symmetry of the circular wafer mid-surface ω\omega. (a) Red boundaries: dashed lines denote symmetry edges Γx,Γy\Gamma_x,\Gamma_y (zero normal slope); solid arc denotes the outer boundary ΓR\Gamma_R (Dirichlet conditions). The striker region ωS\omega_S is shown near the origin. (b) The corresponding finite-element mesh (42,449 triangular elements) used in the DOLFINx computation.

Plate residual. Starting from ∫ωqMαβ,αβ v dω=0\int_{\omega_q} M_{\alpha\beta,\alpha\beta}\,v\,\mathrm{d}\omega=0, two integrations by parts give

Rw(wn+1,pn+1;v)=∫ωqM(wn+1):∇2v dω+∫ωq(pn+1−qextn+1)v dω+∫ωq[−fS(wn+1)−fB(wn+1)−fC(wn+1)+fC(wn+1)wn+1]v dω(40)\mathcal{R}_w(w^{n+1},p^{n+1};v)=\int_{\omega_q}\mathbf{M}(w^{n+1}):\nabla^2v\,\mathrm{d}\omega+\int_{\omega_q}(p^{n+1}-q_{\mathrm{ext}}^{n+1})v\,\mathrm{d}\omega+\int_{\omega_q}\left[-f_S(w^{n+1})-f_B(w^{n+1})-f^C(w^{n+1})+f_C(w^{n+1})w^{n+1}\right]v\,\mathrm{d}\omega \tag*{(40)}

where all boundary terms vanish on ΓR\Gamma_R (essential BC) and on Γx∪Γy\Gamma_x\cup\Gamma_y (symmetry, Mnn=0M_{nn}=0, ∂nw=0\partial_nw=0).

Pressure residual. Multiplying (39) by qq and integrating by parts:

Rp(wn+1,pn+1;q)=∫ωq12μ(wn+1−wn)q dω+Δtn∫ωqh^(wn+1)3∇pn+1⋅∇q dω(41)\mathcal{R}_p(w^{n+1},p^{n+1};q)=\int_{\omega_q}12\mu\left(w^{n+1}-w^n\right)q\,\mathrm{d}\omega+\Delta t_n\int_{\omega_q}\hat{h}(w^{n+1})^3\nabla p^{n+1}\cdot\nabla q\,\mathrm{d}\omega \tag*{(41)}

where the boundary flux term vanishes on ΓR\Gamma_R (pn+1=p∞p^{n+1}=p_\infty, Dirichlet) and on the symmetry edges (zero normal flux by symmetry).

C0 interior-penalty spatial discretization

Since the plate weak form (40) involves second derivatives of ww, it requires H2H^2-conforming elements. Instead, we use the C0C^0 interior-penalty (C0IP) method [29], which employs standard C0C^0-continuous Lagrange elements and enforces inter-element C1C^1-continuity weakly through penalty terms on interior facets.

Finite element spaces. Let Th\mathcal{T}_h be a conforming triangulation of ωq\omega_q and define the quadratic Lagrange spaces

Vh={ϕh∈C0(ωq‾):ϕh∣K∈P2(K) ∀K∈Th},(42)V_h=\left\{\phi_h\in C^0(\overline{\omega_q}):\left.\phi_h\right|_K\in\mathbb{P}_2(K)\ \forall K\in\mathcal{T}_h\right\}, \tag*{(42)}
Qh={ψh∈C0(ωq‾):ψh∣K∈P2(K) ∀K∈Th}.(43)Q_h=\left\{\psi_h\in C^0(\overline{\omega_q}):\left.\psi_h\right|_K\in\mathbb{P}_2(K)\ \forall K\in\mathcal{T}_h\right\}. \tag*{(43)}

and the mixed space Wh=Vh×QhW_h = V_h \times Q_h.

Interior-penalty terms. Let Fhi\mathcal{F}_h^i denote the set of interior facets of Th\mathcal{T}_h, and for any facet e∈Fhie \in\mathcal{F}_h^i let ne\mathbf{n}_e be a fixed unit normal, heh_e the facet diameter, and define the average { ⁣{⋅} ⁣}\{\!\{\cdot\}\!\} and jump [⋅][\cdot] operators. The normal bending moment and normal slope on a facet are

Mnn(wh)=ne⋅M(wh)ne,[∂nwh]=[∇wh⋅ne].(44)M_{nn}(w_h) = \mathbf{n}_e \cdot\mathbf{M}(w_h)\mathbf{n}_e,\qquad[\partial_n w_h] = [\nabla w_h \cdot\mathbf{n}_e]. \tag*{(44)}

The C0IP bending residual, replacing the continuous form (40), reads

Rwh(whn+1,phn+1;vh)=∫ωqM(whn+1):∇2vh dω−∫Fhi{ ⁣{Mnn(whn+1)} ⁣}[∂nvh] dΓ−∫Fhi{ ⁣{Mnn(vh)} ⁣}[∂nwhn+1] dΓ+∫FhiαipDhe[∂nwhn+1][∂nvh] dΓ+∫ωq(phn+1−qextn+1)vh dω+∫ωq[−fS(whn+1)−fB(whn+1)−fC(whn+1)+fC(whn+1)whn+1]vh dω,(45)\begin{aligned} \mathcal{R}_w^h(w_h^{n+1},p_h^{n+1};v_h)={}&\int_{\omega_q}\mathbf{M}(w_h^{n+1}):\nabla^2v_h\,\mathrm{d}\omega\\ &-\int_{\mathcal{F}_h^i}\{\!\{M_{nn}(w_h^{n+1})\}\!\}[\partial_n v_h]\,\mathrm{d}\Gamma-\int_{\mathcal{F}_h^i}\{\!\{M_{nn}(v_h)\}\!\}[\partial_n w_h^{n+1}]\,\mathrm{d}\Gamma\\ &+\int_{\mathcal{F}_h^i}\frac{\alpha_{\mathrm{ip}}D}{h_e}[\partial_n w_h^{n+1}][\partial_n v_h]\,\mathrm{d}\Gamma+\int_{\omega_q}(p_h^{n+1}-q_{\mathrm{ext}}^{n+1})v_h\,\mathrm{d}\omega\\ &+\int_{\omega_q}\left[-f_S(w_h^{n+1})-f_B(w_h^{n+1})-f^C(w_h^{n+1})+f_C(w_h^{n+1})w_h^{n+1}\right]v_h\,\mathrm{d}\omega, \tag*{(45)} \end{aligned}

where αip>0\alpha_{\mathrm{ip}}>0 is the interior-penalty parameter.

On the symmetry boundaries Γs∈{Γx,Γy}\Gamma_s\in\{\Gamma_x,\Gamma_y\}, the condition ∂nw=0\partial_n w=0 (cf. (26)) is enforced weakly via Nitsche-type terms consistent with the C0IP formulation:

Rsymh(whn+1;vh)=∑Γs∈{Γx,Γy}[−∫ΓsMnn(whn+1)∂nvh dΓ−∫ΓsMnn(vh)∂nwhn+1 dΓ+∫ΓsαipDhe∂nwhn+1∂nvh dΓ].(46)\mathcal{R}_{\mathrm{sym}}^h(w_h^{n+1};v_h)=\sum_{\Gamma_s\in\{\Gamma_x,\Gamma_y\}}\left[-\int_{\Gamma_s}M_{nn}(w_h^{n+1})\partial_n v_h\,\mathrm{d}\Gamma-\int_{\Gamma_s}M_{nn}(v_h)\partial_n w_h^{n+1}\,\mathrm{d}\Gamma+\int_{\Gamma_s}\frac{\alpha_{\mathrm{ip}}D}{h_e}\partial_n w_h^{n+1}\partial_n v_h\,\mathrm{d}\Gamma\right]. \tag*{(46)}

The discrete pressure residual is identical to the continuous form (41) restricted to QhQ_h:

Rph(whn+1,phn+1;qh)=∫ωq12μ(whn+1−whn)qh dω+Δtn∫ωqh^(whn+1)3∇phn+1⋅∇qh dω.(47)\mathcal{R}_p^h(w_h^{n+1},p_h^{n+1};q_h)=\int_{\omega_q}12\mu\left(w_h^{n+1}-w_h^n\right)q_h\,\mathrm{d}\omega+\Delta t_n\int_{\omega_q}\hat{h}(w_h^{n+1})^3\nabla p_h^{n+1}\cdot\nabla q_h\,\mathrm{d}\omega. \tag*{(47)}

Monolithic nonlinear system and Newton solver

The fully discrete monolithic residual is assembled as

Rh(whn+1,phn+1;vh,qh)=Rwh(whn+1,phn+1;vh)+Rsymh(whn+1;vh)+Rph(whn+1,phn+1;qh).(48)\mathcal{R}^h(w_h^{n+1},p_h^{n+1};v_h,q_h)=\mathcal{R}_w^h(w_h^{n+1},p_h^{n+1};v_h)+\mathcal{R}_{\mathrm{sym}}^h(w_h^{n+1};v_h)+\mathcal{R}_p^h(w_h^{n+1},p_h^{n+1};q_h). \tag*{(48)}

At each time step tn+1t_{n+1}, we seek (whn+1,phn+1)∈Wh(w_h^{n+1},p_h^{n+1})\in W_h such that

Rh(whn+1,phn+1;vh,qh)=0∀ (vh,qh)∈Wh0,(49)\mathcal{R}^h(w_h^{n+1},p_h^{n+1};v_h,q_h)=0\qquad\forall\,(v_h,q_h)\in W_h^0, \tag*{(49)}

where Wh0⊂WhW_h^0\subset W_h incorporates homogeneous essential boundary conditions (vh=0v_h=0 on ΓR\Gamma_R, qh=0q_h=0 on Γvent\Gamma_{\mathrm{vent}}). The essential (Dirichlet) conditions enforced are:

  • phn+1=0p_h^{n+1}=0 on the vented outer boundary ΓR\Gamma_R (gauge pressure),

  • whn+1=wedge(tn+1)w_h^{n+1}=w_{\mathrm{edge}}(t_{n+1}) on ΓR\Gamma_R,

  • whn+1=wcenter(tn+1)w_h^{n+1}=w_{\mathrm{center}}(t_{n+1}) at the striker/pin region.

Due to the nonlinear loading terms fB(w)f_B(w), fC(w)f_C(w), ftc(w)f_{\mathrm{tc}}(w) and the cubic gap dependence h^(w)3\hat{h}(w)^3 in the Reynolds equation, the discrete problem (49) is nonlinear and is solved by Newton’s method. Denoting the algebraic unknown vector U=(ωh,ph)⊤\mathbf{U}=(\boldsymbol{\omega}_h,\mathbf{p}_h)^{\top} and the algebraic residual vector R(U)\mathbf{R}(\mathbf{U}), the Newton iteration reads

J(U(k))ΔU(k)=−R(U(k)),U(k+1)=U(k)+ΔU(k),(50)\mathbf{J}(\mathbf{U}^{(k)})\Delta\mathbf{U}^{(k)}=-\mathbf{R}(\mathbf{U}^{(k)}),\qquad\mathbf{U}^{(k+1)}=\mathbf{U}^{(k)}+\Delta\mathbf{U}^{(k)}, \tag*{(50)}

where the Jacobian J=∂R/∂U\mathbf{J}=\partial\mathbf{R}/\partial\mathbf{U} is computed automatically via UFL’s [21] symbolic differentiation capabilities in FEniCSx.

The complete solution procedure is summarized in Algorithm 1.

Algorithm 1 Monolithic coupled wafer-wafer bonding simulation in FEniCSx

  1. // -- Mesh, FE Spaces & Measures --

  2. Load Gmsh model via gmsh.model_to_mesh →\to mesh Th\mathcal{T}_h; build connectivity

  3. hip←∣ωq∣/Ncellsh_{\mathrm{ip}} \leftarrow\sqrt{\lvert\omega_q\rvert/N_{\mathrm{cells}}} \hfill IP penalty scale

  4. Wh←W_h \leftarrow functionspace(msh, mixed_element([P2, P2]))

  5. (wh,ph), (vh,qh), (whn,phn)←(w_h,p_h),\ (v_h,q_h),\ (w_h^n,p_h^n) \leftarrow split of trial, test, prev. step Functions

  6. dx, dS ←\leftarrow cell/interior-facet measures; ds_sym ←\leftarrow tagged symmetry facets via meshtags

  7. // -- UFL Variational Forms --

  8. ▹\triangleright Plate residual Rhw\mathcal{R}_h^w (45): F1 ←\leftarrow inner(M,hess)*dx −- avg(M_nn)*jump*dS ++ penalty ++ loads

  9. ▹\triangleright Nitsche symmetry Rsymw\mathcal{R}_{\mathrm{sym}}^w (46): F1 += ∑Γs\sum_{\Gamma_s} [-M_nn*dn_v + αD/h*dn_w*dn_v]*ds_sym

  10. ▹\triangleright Pressure residual Rhp\mathcal{R}_h^p (47): F2 ←\leftarrow 12μ(w-wn)*q*dx + Δt* h^3 *dot(grad(p),grad(q))*dx

  11. residual ←\leftarrow F1+F2; jacobian ←\leftarrow derivative(residual, w_p, TrialFunction) [21]

  12. // -- Boundary Conditions & Solver --

  13. bcs ←\leftarrow [dirichletbc(p=0, Γ_R), dirichletbc(w=w_edge, Γ_R), dirichletbc(w=w_center, ω_S)]

  14. problem ←\leftarrow NonlinearProblem(residual, w_p, bcs, J=jacobian); PETSc LU/MUMPS

  15. Open VTXWriter for w,pw,p output (BP4)

  16. // -- Time-Stepping Loop --

  17. for n=0,1,…,N−1n=0,1,\ldots,N-1 do

  18. tn+1←tn+Δtnt_{n+1}\leftarrow t_n+\Delta t_n; update dt.value, disp_edge.value, disp_center.value from CSV

  19. Solve: problem.solve() \hfill assembles R,J\mathbf{R},\mathbf{J}; Newton iteration (50)

  20. w_p_n.x.array[:] = w_p.x.array[:]; w_sol.interpolate(w_sub)

  21. w_vtx.write(t_{n+1}); p_vtx.write(t_{n+1}) \hfill write output

  22. end for

  23. w_vtx.close(); p_vtx.close(); save radial profiles via np.save

  24. return (whN,phN)(w_h^N,p_h^N)

Results

This section presents the numerical results for the wafer-bonding model; the solution workflow is outlined in Algorithm 1. We first validate the implementation against experimental top-wafer displacement measurements and illustrate the resulting quarter-symmetric displacement field (Figure 4). We then verify the force balance at the bond front by comparing the Reynolds air pressure with the computed contact reaction (§ 4.2). Next, we assess how the initial bond gap affects air-pressure build-up and the bond-front propagation speed (§ 4.3), and we conduct a parametric study on the role of air viscosity. Finally, we examine how the interfacial energy influences the bonding speed.

Image not extracted

Figure 4. Snapshots of the displacement field ww at nine time instances during bonding for h0=70 μmh_0 = 70\,\mu\mathrm{m}. The out-of-plane displacement is warped by a factor of 500 to enhance the visualization. The bottom wafer is shown in gray, and the color scale indicates the physical displacement of the top wafer.

Terminology note: Here “bond speed” denotes the radial gap-closure rate r˙b=drb/dt\dot{r}_b = \mathrm{d}r_b/\mathrm{d}t, where rbr_b is the radius from the center-connected region satisfying the bonding criterion h≤δh \le\delta (region already bonded). It is a post-processed radial propagation rate obtained from the out-of-plane Kirchhoff–Love plate solution, rather than a material-point velocity.

Table 2 summarizes the input parameters common to all simulations presented in this section. Parameters marked with † were calibrated as described in § 4.1.

ParameterSymbolValueSource
Wafer radiusRR150 mmStandard 12" wafer
Wafer thicknesstwt_w0.775 mmStandard 12" wafer
Young’s modulusEE148 GPaStandard Si
Poisson’s ratioν\nu0.27Standard Si
Air viscosityμ\mu1.86×10−51.86 \times10^{-5} Pa·sStandard (20°C)
Pin (striker) radiusRS/RpinR_{\mathrm{S}}/R_{\mathrm{pin}}5.0 mmAssumed
Initial bond gapsh0h_030, 70, 100 μ\mum[11]
Regularization parameterβ\beta0.0075Sec. 3.1; App. B
Bonding traction†fBEf_{\mathrm{BE}}0.03 MPaCalibrated
Bonding threshold†δ\delta0.001 mmCalibrated
Interfacial energy†Γ=fBEδ\Gamma= f_{\mathrm{BE}}\delta0.03 J/m2^2Calibrated
Striker pressure†fSf_{\mathrm{S}}0.028 MPaCalibrated
Displacement locationrmid-regionr_{\mathrm{mid\text{-}region}}75 mmAssumed

Table 2. Input parameters for the wafer bonding simulations.

Validation

To validate the implementation, we performed three simulations corresponding to the bond gaps (a) h0=30 μmh_0 = 30\,\mu\mathrm{m}, (b) h0=70 μmh_0 = 70\,\mu\mathrm{m}, and (c) h0=100 μmh_0 = 100\,\mu\mathrm{m}. In each case, Dirichlet boundary condition w=wΓw = w_{\Gamma} was prescribed using the displacement history reported in Figure 3 of [11]. The resulting displacement time history at the mid-region3 of the top wafer was then compared with the experimental measurements. Figure 3 shows that the model reproduces the measured response with good qualitative agreement and reasonable quantitative accuracy. Small discrepancies are expected because several inputs required for an exact match are not fully specified in [11] (e.g., interfacial energy and the precise probe location); these were therefore inferred through a calibration procedure in this work.

Image not extracted

Figure 3. Top wafer displacement w(r,t)w(r,t) versus time at a fixed radial location r=mid-regionr = \mathit{mid\text{-}region} for h0=30h_0 = 30, 70, and 100 µm. Solid lines show the finite-element solution; dashed lines show the experimental measurement.

The parameters not specified in [11] are the striker pressure fSf_{\mathrm{S}}, the bonding traction magnitude fBEf_{\mathrm{BE}} (Eq. (3)), the bonding activation threshold δ\delta, and the radial location of the displacement sensor. The sensor location was assumed to be r=75r = 75 mm based on the reported experimental setup. The striker pressure, the bonding traction and threshold were calibrated by minimizing the discrepancy between the simulated and measured displacement time histories at the sensor location. The calibrated values are fBE=0.03f_{\mathrm{BE}} = 0.03 MPa and δ=0.001\delta= 0.001 mm (=1= 1 µm). The product Γ=fBEδ=0.03\Gamma= f_{\mathrm{BE}} \delta= 0.03 J/m2^{2} corresponds to the interfacial bonding energy, which falls within the range of 20–100 mJ/m2^{2} reported for room-temperature hydrophilic silicon wafer bonding [3, 7]. Furthermore, since the wafer release condition was not reported in [11], the measured edge displacement time history was instead prescribed as a displacement boundary condition on the edge ΓR\Gamma_{\mathrm{R}}. The striker pressure, applied uniformly over the central region r≤RSr \leq R_{\mathrm{S}}, was calibrated by matching the measured wafer-center displacement history, giving fS=0.028f_{\mathrm{S}} = 0.028 MPa.

The effect of air pressure is evident in both the experiments and the simulations. In contrast to what is expected for bonding in the absence of air (pure-plate-bending), in the presence of air, the largest initial bond gap reaches complete bonding earliest, whereas the smallest gap reaches it last. This ordering is also evident in the bonded-radius evolution discussed in § 4.3.

Snapshots of the simulated out-of-plane displacement field are shown as 2D contour plots at selected time instants in Figure 4.

Reynold's pressure acting as a contact force

As discussed in § 2.5, we set the bottom-wafer contact tractions to zero, fC(w)=0f_C(w)=0, because the air-film pressure obtained from the Reynolds solution, pp, is sufficient to enforce non-penetration contact. This can be seen from Eq. (37): the Reynolds equation is solved over the full computational domain ω\omega, including regions that are already bonded. In the bonded region the Reynolds pressure acts as a reaction pressure, i.e., an effective contact traction that prevents interpenetration of the wafers. This is because of the inherent mathematical structure of the Reynold’s equation in which hh in Eq. (37) cannot be negative. Figure 5 confirms this interpretation by showing that the Reynolds pressure behind the bond front balances the applied striker pressure. So, at the bond front, the total Reynold’s pressure pp is:

Image not extracted

Figure 5. Displacement w(r)w(r) (left) and pressure p(r)p(r) (right) at the final time t=50t=50 s for h0=70h_0=70 μ\mum. The average pressure pˉ\bar{p} is computed by integrating the Reynolds pressure over r≤RS=Rpinr\leq R_S=R_{\mathrm{pin}}; the resulting average pressure is pˉ=fS=0.028\bar{p}=f_S=0.028 MPa.

p(xˉ,t)=pˉ(xˉ,t)+pair(xˉ,t),(51)p(\bar{x},t)=\bar{p}(\bar{x},t)+p_{\mathrm{air}}(\bar{x},t), \tag*{(51)}

where, pairp_{\mathrm{air}} is the actual pressure due to the air between the two wafers in the unbonded region and pˉ\bar{p} is the contact pressure.

We note that the Reynolds pressure computed in the bonded region (where the effective gap approaches h^→βh0\hat{h}\to\beta h_0) does not represent physical air pressure, as the air has been expelled from bonded areas. Rather, this pressure serves as an effective contact reaction that enforces the non-penetration constraint, analogous to a penalty-regularized contact formulation. This interpretation is validated by the force balance shown in Figure 5. The regularization parameter β\beta (cf. § 3.1) is essential for this contact-reaction interpretation: by maintaining a minimum effective gap, it prevents the pressure singularity that would occur as h→0h\to0 in the unregularized Reynolds equation while allowing the Reynolds equation to return a finite, well-defined pressure in bonded regions.

Effect of bond gap on bond speed

In this section, we examine the bonding speed for the three bond gaps considered in the validation exercise in § 4.1. As shown in Figure 6, the response is nonlinear and non-monotonic. Unlike bonding in the absence of air (pure plate bending), the largest bond gap reaches complete bonding earlier than the smallest gap; see Figure 6(a). This ordering is driven by the air-film pressure pairp_{\mathrm{air}} in the unbonded region: smaller gaps generate higher pairp_{\mathrm{air}}, which increases the resistance to bond-front advance as in demonstrated in Figure 10b. The instantaneous bond-front speed remains nonlinear and is not ordered monotonically for all times, as illustrated in Figure 6(b) and Figure 6(c). This behavior is expected, since the bond-front speed and the evolution of air pressure are governed by Reynolds’ equation. Additional validation plots for the spatio-temporal displacement fields and air-pressure diagnostics are provided in Appendix A; see Figure 9 and Figure 10.

Image not extracted

Figure 6. Effect of the initial bond gap for h0=30h_0 = 30, 7070, and 100 μm100\,\mu\mathrm{m}: (a) bonded radius versus time, (b) bond speed versus time, and (c) bond speed versus radial position. Before bond initiation, rb=0r_b = 0 and r˙b=0\dot{r}_b = 0.

Figure 9. Spatio-temporal evolution of the radial displacement field w(r,t)w(r,t) for three bond gaps.

Figure 10. Air-pressure diagnostics for h0=30h_0 = 30, 7070, and 100 μm100\,\mu\mathrm{m} for the result reported in § 4.3: (a) radial profiles pair(r)p_{\mathrm{air}}(r) in the unbonded region at t=9 st = 9\,\mathrm{s}; and (b) profiles at time t=tct = t_c when rb(tc)=65 mmr_b(t_c) = 65\,\mathrm{mm}. The inset in (b) magnifies the bond-front region 65≤r≤7065 \le r \le70.

Effect of Air Viscosity on air pressure and Bond Speed

Varying the air viscosity modifies the Reynolds pressure field and, consequently, the bond-front propagation rate. Figure 7 summarises the post-processed diagnostics used to quantify this effect: bond speed versus time, the spatial variation of bond speed, and the temporal evolution of the bonded radius. For the same time instant, lower viscosity yields a faster bond front and a larger bonded area.

Image not extracted

Figure 7. Effect of air viscosity for h0=70 μmh_0 = 70\,\mu\mathrm{m} and μ/μ0=0.25\mu/\mu_0 = 0.25, 0.50, 0.75, and 1.00: (a) bond speed over time, (b) bond speed versus radius, and (c) bonded-radius evolution, where μ0=1.86×10−5 Pa⋅s\mu_0 = 1.86 \times10^{-5}\,\mathrm{Pa}\cdot\mathrm{s} is the reference air viscosity.

Effect of Interfacial Energy on bond speed

Varying the interfacial energy Γ\Gamma changes the thermodynamic driving force for bonding, which in turn affects the bond speed. Figure 8 presents the diagnostics that capture this sensitivity: the bond speed as a function of time, the spatial distribution of the bond speed, and the temporal evolution of the bonded radius.

Image not extracted

Figure 8. Effect of interfacial energy for h0=70 μmh_0 = 70\,\mathrm{\mu m} and fBE=0.005f_{\mathrm{BE}} = 0.005, 0.010, 0.020, and 0.030 MPa, corresponding to Γ=0.005\Gamma= 0.005, 0.010, 0.020, and 0.030 J m−2\mathrm{m}^{-2}: (a) bond speed over time, (b) bond speed versus radius, and (c) bonded-radius evolution.

Conclusion and Future Work

Conclusions

This work developed a mathematically consistent reduced-order model for wafer-to-wafer bonding by coupling a Kirchhoff–Love plate description of wafer bending to a Reynolds lubrication model for the entrapped air film. Starting from three-dimensional linear elasticity, the plate equation was obtained in the thin-wafer limit and combined with a pressure evolution law in which the wafer deflection directly modulates the local gap height and flow resistance. The resulting nonlinear, strongly coupled system was discretized in a monolithic manner and implemented in the FEniCSx framework using a C0C^0 interior-penalty formulation for the fourth-order plate operator together with standard continuous Galerkin elements for the pressure field. The numerical studies demonstrated consistent force balance in the contact region and produced bonding-front dynamics in agreement with experimentally measured wafer-displacement histories, thereby providing a predictive and computationally efficient tool for assessing process sensitivities.

A key advantage of the present formulation is that the coupling structure is explicit: the air pressure enters the plate equation as a distributed transverse load, while the plate deflection enters the Reynolds equation through the evolving gap h=h0+wh = h_0 + w. This structure enables systematic extensions of the physics without altering the overall solution strategy.

Extension to anisotropic flexural response

In the current isotropic plate model, the bending stiffness enters through the scalar flexural rigidity D=Etw3/[12(1−ν2)]D = Et_w^3/[12(1-\nu^2)]. For crystalline wafers and layered stacks, however, the effective bending response can be direction-dependent. Within the same Kirchhoff–Love framework, anisotropy can be incorporated by replacing the scalar stiffness with a bending-stiffness tensor. In practice, one introduces the bending moment–curvature relation

M=D κ,(52)M = \mathbb{D}\,\kappa, \tag*{(52)}

where MM and κ=∇2w\kappa= \nabla^2 w are the bending-moment and curvature tensors, respectively, and D\mathbb{D} is a symmetric fourth-order bending stiffness (often expressed, under plane-stress reduction, as a 3×33 \times3 matrix in Voigt notation). The plate equilibrium then becomes

∇⋅∇⋅(D κ)−qext+pair−fB(w)−fc(w)=0,(53)\nabla\cdot\nabla\cdot(\mathbb{D}\,\kappa) - q_{\mathrm{ext}} + p_{\mathrm{air}} - f_{\mathrm{B}}(w) - f_{\mathrm{c}}(w) = 0, \tag*{(53)}

which reduces to DΔ2wD \Delta^2 w when D\mathbb{D} corresponds to isotropic bending. Because the Reynolds equation and the coupling through hh remain unchanged, the anisotropic extension primarily modifies the plate operator and the constitutive evaluation of bending moments, and can be integrated into the present monolithic Newton framework with minimal changes.

Future work

Several extensions are planned. First, we will incorporate anisotropic and heterogeneous bending stiffness to represent crystallographic effects, device layers, and patterned stacks, enabling more accurate prediction of bonding kinetics for modern wafer architectures. Second, we will extend the reduced two-dimensional plate–Reynolds formulation to a geometrically richer shell model. A shell description will allow simultaneous prediction of (i) bond-front propagation and pressure evolution in the interfacial air film and (ii) in-plane deformations and post-bond overlays, including wafer stretch, shear, and in-plane distortion. This, in turn, will enable prediction of post-bond wafer shapes and residual distortions (e.g., global bow/warp and in-plane misfit patterns) that are critical for overlay, alignment, and downstream process yield. Finally, we will investigate the effect on bonding due to incoming wafer geometries.

Additional validation results

This appendix section collects additional plots supporting the validation study in § 4.1. Figure 9 presents the full spatio-temporal displacement field for each initial gap, from which the bond-front trajectories reported in Figure 6(a) are extracted. Figure 10 presents the radial air-pressure profiles that explain the effect of bond gap on bond speed as discussed in § 4.3.

Sensitivity to regularization parameter

This appendix presents the sensitivity of the simulation results to the regularization parameter β\beta introduced in § 3.1. Simulations were performed for h0=70 μmh_0 = 70\,\mu\mathrm{m} with β∈{0.0025,0.005,0.0075,0.01}\beta\in\{0.0025, 0.005, 0.0075, 0.01\}. Figure 11 demonstrates that the displacement and pressure are approximately insensitive to β\beta within this range, confirming that the choice β=0.0075\beta= 0.0075 used throughout this work does not materially affect the reported results.

Image not extracted

Figure 11. Effect of regularization parameter β\beta on successive difference norms for h0=70 μmh_0 = 70\,\mu\mathrm{m}: (a) normalized deflection difference ∥wβi−wβi−1∥L2\lVert w_{\beta_i} - w_{\beta_{i-1}} \rVert_{L^2} versus time, and (b) normalized pressure difference ∥pβi−pβi−1∥L2\lVert p_{\beta_i} - p_{\beta_{i-1}} \rVert_{L^2} versus time, for consecutive pairs from β∈{0.0025,0.005,0.0075,0.01}\beta\in\{0.0025, 0.005, 0.0075, 0.01\}.

References

  1. [1]Shuyu Bao, Yue Wang, Khaw Lina, Li Zhang, Bing Wang, Wardhana Aji Sasangka, Kenneth Eng Kian Lee, Soo Jin Chua, Jurgen Michel, Eugene Fitzgerald, Chuan Seng Tan, and Kwang Hong Lee. A review of silicon-based wafer bonding processes, an approach to realize the monolithic integration of Si-CMOS and III–V-on-Si wafers. Journal of Semiconductors, 42 (2):023106, 2021. doi: 10.1088/1674-4926/42/2/023106.DOI
  2. [2]H. Moriceau, F. Rieutord, F. Fournel, Y. Le Tiec, L. Di Cioccio, C. Morales, A. M. Charvet, and C. Deguet. Overview of recent direct wafer bonding advances and applications. Advances in Natural Sciences: Nanoscience and Nanotechnology, 1(4):043004, 2011. doi: 10.1088/2043-6262/1/4/043004.DOI
  3. [3]U. Gösele and Q.-Y. Tong. Semiconductor wafer bonding. Annual Review of Materials Science, 28:215–241, 1998. doi: 10.1146/annurev.matsci.28.1.215.DOI
  4. [4]L. Wang, G. Fountain, B. Lee, G. Gao, C. Uzoh, S. McGrath, P. Enquist, S. Arkalgud, and L. Mirkarimi. Direct bond interconnect (DBI) for fine-pitch bonding in 3D and 2.5D integrated circuits. 2023 IEEE 73rd Electronic Components and Technology Conference (ECTC), pages 1–8, 2023. doi: 10.37665/ppobdcm64121.
  5. [5]K. T. Turner and S. M. Spearing. Modeling of direct wafer bonding: Effect of wafer bow and etch patterns. Journal of Applied Physics, 92(12):7658–7666, 2002. doi: 10.1063/1.1521792.DOI
  6. [6]F. Rieutord, B. Bataillou, and H. Moriceau. Dynamics of a bonding front. Physical Review Letters, 94:236101, 2005. doi: 10.1103/PhysRevLett.94.236101.DOI
  7. [7]V. Larrey, G. Mauguen, F. Fournel, D. Radisson, F. Rieutord, C. Morales, C. Bridoux, and H. Moriceau. Adhesion energy and bonding wave velocity measurements. ECS Transactions, 75(9):145–152, 2016. doi: 10.1149/07509.0145ecst.DOI
  8. [8]E. Navarro, Y. Bréchet, R. Moreau, T. Pardoen, J.-P. Raskin, O. Barthelemy, and I. Radu. Direct silicon bonding dynamics: a coupled fluid/structure analysis. Applied Physics Letters, 103:034104, 2013. doi: 10.1063/1.4813312.DOI
  9. [9]D. Radisson, F. Fournel, and E. Charlaix. Modelling of the direct bonding wave. Microsystem Technologies, 21:969–971, 2015. doi: 10.1007/s00542-015-2445-3.DOI
  10. [10]Utkarsh Jain. Bond front velocity in lubrication-mediated bonding of flexible substrates. Proceedings of the Royal Society A, 482:20260071, 2026. doi: 10.1098/rspa.2026.0071.DOI
  11. [11]Nathan Ip, Nima Nejadsadeghi, Carlos Fonseca, Norifumi Kohama, and Kimio Motoda. Multi-physics simulation of wafer-to-wafer bonding dynamics. In 2022 IEEE 72nd Electronic Components and Technology Conference (ECTC), pages 502–506. IEEE, 2022. doi: 10.1109/ECTC51906.2022.00085.DOI
  12. [12]Kyeongbin Lim et al. Design and simulation of symmetric wafer-to-wafer bonding compensating a gravity effect. In 2020 IEEE 70th Electronic Components and Technology Conference (ECTC), pages 1480–1485. IEEE, 2020. doi: 10.1109/ECTC32862.2020.00234.DOI
  13. [13]Feixiang Tang, Siyu He, Yuhan Li, Wenjin Liu, Fang Dong, and Sheng Liu. Numerical and mechanical analysis of direct wafer bonding considering non-uniform impurity particle distributions. Microsystems & Nanoengineering, 11:151, 2025. doi: 10.1038/s41378-025-00994-4.DOI
  14. [14]Seyed Amir Fouad Farshchi Yazdi, Matteo Garavaglia, Aldo Ghisi, and Alberto Corigliano. An experimental and numerical study on glass frit wafer-to-wafer bonding. Micromachines, 14(1):165, 2023. doi: 10.3390/mi14010165.
  15. [15]Yan Wen Tsau, Joke De Messemaeker, Abdellah Salahouelhadj, Mario Gonzalez, Liesbeth Witters, Boyao Zhang, Marc Seefeldt, Eric Beyne, and Ingrid De Wolf. Simulation of Cu pad expansion in wafer-to-wafer Cu/SiCN hybrid bonding. Microelectronics Reliability, 138:114716, 2022. doi: 10.1016/j.microrel.2022.114716.DOI
  16. [16]Igor A Baratta, Joseph P Dean, Jørgen S Dokken, Michal Habera, Jack Hale, Chris N Richardson, Marie E Rognes, Matthew W Scroggs, Nathan Sime, and Garth N Wells. DOLFINx: the next generation FEniCS problem solving environment. Zenodo, 2023. doi: 10.5281/zenodo.10447666. URL https://doi.org/10.5281/zenodo.10447666. Preprint.
  17. [17]Kamalendu Ghosh and Bhavesh Shrimali. A hybrid finite element implementation of two-potential constitutive model of dielectric elastomers. Finite Elements in Analysis and Design, 247:104348, 2025. doi: 10.1016/j.finel.2025.104348.DOI
  18. [18]Bhavesh Shrimali, Matteo Pezzulla, Samuel Poincloux, Pedro M Reis, and Oscar Lopez-Pamies. The remarkable bending properties of perforated plates. Journal of the Mechanics and Physics of Solids, 154:104514, 2021. doi: 10.1016/j.jmps.2021.104514.DOI
  19. [19]Matthew W Scroggs, Jørgen S Dokken, Chris N Richardson, and Garth N Wells. Construction of arbitrary order finite element degree-of-freedom maps on polygonal and polyhedral cell meshes. ACM Transactions on Mathematical Software, 48(2):1–23, 2022.DOI
  20. [20]Matthew W Scroggs, Igor A Baratta, Chris N Richardson, and Garth N Wells. Basics: a runtime finite element basis evaluation library. Journal of Open Source Software, 7(73):3982, 2022.DOI
  21. [21]Martin S Alnæs, Anders Logg, Kristian B Ølgaard, Marie E Rognes, and Garth N Wells. Unified form language: A domain-specific language for weak formulations of partial differential equations. ACM Transactions on Mathematical Software, 40(2):1–37, 2014.
  22. [22]Kamalendu Ghosh, Bhavesh Shrimali, Aditya Kumar, and Oscar Lopez-Pamies. The nonlinear viscoelastic response of suspensions of rigid inclusions in rubber: I—Gaussian rubber with constant viscosity. Journal of the Mechanics and Physics of Solids, 154:104544, 2021. doi: 10.1016/j.jmps.2021.104544.DOI
  23. [23]Bhavesh Shrimali, Kamalendu Ghosh, and Oscar Lopez-Pamies. The nonlinear viscoelastic response of suspensions of vacuous bubbles in rubber: I—Gaussian rubber with constant viscosity. Journal of Elasticity, 153(4):479–508, 2023. doi: 10.1007/s10659-021-09868-y.
  24. [24]Kamalendu Ghosh and Oscar Lopez-Pamies. On the two-potential constitutive modeling of dielectric elastomers. Meccanica, 56(6):1505–1521, 2021. doi: 10.1007/s11012-020-01179-1.DOI
  25. [25]Bernard J. Hamrock, Steven R. Schmid, and Bo O. Jacobson. Fundamentals of Fluid Film Lubrication. Marcel Dekker, New York, 2nd edition, 2004.DOI
  26. [26]Albert Burgdorfer. The Influence of the Molecular Mean Free Path on the Performance of Hydrodynamic Gas Lubricated Bearings. Journal of Basic Engineering, 81(1):94–98, 1959. doi: 10.1115/1.4008375.DOI
  27. [27]Shigehisa Fukui and Raj Kaneko. Analysis of Ultra-Thin Gas Film Lubrication Based on Linearized Boltzmann Equation: First Report—Derivation of a Generalized Lubrication Equation Including Thermal Creep Flow. Journal of Tribology, 110(2):253–261, 1988. doi: 10.1115/1.3261594.DOI
  28. [28]Etienne Navarro. Direct Wafer Bonding Dynamics. PhD thesis, Université de Grenoble, 2014. HAL Id: tel-01048574.DOI
  29. [29]Tom Gustafsson, Rolf Stenberg, and Juha Videman. On Nitsche’s method for elastic contact problems. SIAM Journal on Scientific Computing, 42(2):B425–B446, 2020. doi: 10.1137/19M1246869.DOI
  30. [30]C. Teichert, J. F. MacKay, D. E. Savage, M. G. Lagally, M. Brohl, and P. Wagner. Comparison of surface roughness of polished silicon wafers measured by light scattering topography, soft-x-ray scattering, and atomic-force microscopy. Applied Physics Letters, 66(18):2346–2348, 1995. doi: 10.1063/1.113978.DOI
  31. [31]Guy Bayada and Michèle Chambat. The Transition Between the Stokes Equations and the Reynolds Equation: A Mathematical Proof. Applied Mathematics and Optimization, 14(1):73–93, 1986. doi: 10.1007/BF01442229.DOI

Paper details

vlnc:W_wjhewyfr
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 10

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents